Tom Ryerson NOAA ESRL Chemical Sciences Division CalNex 2010: NOAA perspective Goal of this...

Preview:

Citation preview

Tom RyersonNOAA ESRLChemical Sciences Division

http://esrl.noaa.gov/csd/calnex/scienceplan.pdf

CalNex 2010: NOAA perspective

Goal of this presentation:Summarize study planning

activities and opportunities for collaboration.

WRAP Ozone & NOx in the West meetingNovember 11, 2009

Santa Fe, NM

today:

• Introduction to CalNex 2010

• Introduction to NOAA’s study interests & platforms

• Opportunities for collaboration

Why CalNex?

http://esrl.noaa.gov/csd/calnex/scienceplan.pdf

CalNex 2010: NOAA perspective

WRAP Ozone & NOx in the West meetingNovember 11, 2009

Santa Fe, NM

1) Management and Mitigation Strategies for these two issues are strongly linked.

2) Sources and Processes for these two issues are the same (nearly).

Why California?

http://esrl.noaa.gov/csd/calnex/scienceplan.pdf

CalNex 2010: NOAA perspective

• California has recently taken the lead in beginning to address Climate Change.

• California has traditionally led the nation in addressing Air Quality issues.

• NOAA is developing a new research initiative addressing Air Quality and Climate Change linkages.

Study Rationale

CalNex addresses NOAA’s andCalifornia’s Science Interests

What are NOAA’s interests?http://esrl.noaa.gov/csd/calnex/scienceplan.pdf

CalNex 2010: NOAA perspective

NOAA/ESRL Regional IntensivesNOAA/ESRL Regional Intensives

Chemically and Meteorologically DiverseChemically and Meteorologically Diverse

California

East Texas2000, 2006

Southeast1995, 1999

New England

1993, 2002, 2004,

2008

ITCT 2002

CCOS 2000

What will NOAA contribute to CalNex?

Study Air Quality and Climate Change Issues through a Series of Regional Intensives

NEAQS

SOS

TexAQS

NOAA’s AssetsNOAA’s Assets

NOAA WP-3D

NOAA R/V Ronald H. Brown NOAA LIDAR Aircraft

Collaborate with Others on fielding Ground-based Remote and In Situ Instrumentation

• Aerosol Properties and Radiative effects

• Chemical Transformation and Climate Processes

• Emissions - top down tests

• Transport and Meteorology

Science Issues

Some illustrations

http://esrl.noaa.gov/csd/calnex/scienceplan.pdf

CalNex 2010: NOAA perspective

• Forecast Model Evaluation

Top-down tests of emission inventories:

•Aerosol and ozone precursors,

•Greenhouse gases,

•Black carbon, of particular interest.

Emissions QuantificationEmissions QuantificationImproved inventories are essential for predictive capability Improved inventories are essential for predictive capability

Chemical TransformationChemical TransformationAim to understand on a process level for predictive capability Aim to understand on a process level for predictive capability

Investigate:

•Secondary organic aerosol formation

•NH4NO3 formation

•Photochemistry; ozone formation

•Role of nighttime chemistry

•Halogen activation

Transport and MeteorologyTransport and MeteorologyUnderstanding is critical for characterizing O3 and PM levels Understanding is critical for characterizing O3 and PM levels

Investigate:

•Orographic ventilation and recirculation of pollution in the atmospheric boundary layer

•Inter-regional transport

•Offshore recirculation and marine boundary layer phenomena

•Longer-range transport

Transport of emissions into, within, and out of California

Transport and MeteorologyTransport and MeteorologyUnderstanding is critical for characterizing O3 and PM levels Understanding is critical for characterizing O3 and PM levels

Riley et al., JGR, 113, 10.1029/2007JG000625, 2008

14C in annual grasses3 yr average of maximum daily max 8-hr avg

2005-2007

2010: Investigate transport mechanisms so that we can understand differences in observed patterns of anthropogenic emissions and ozone distributions

Transport of emissions into, within, and out of California

Transport and MeteorologyTransport and MeteorologyUnderstanding is critical for characterizing O3 and PM levels Understanding is critical for characterizing O3 and PM levels

2010: Investigate export of California emissions .

3 yr average of 4th highest daily max 8-hr avg

2004-2006

Aerosol Properties and Radiative EffectsAerosol Properties and Radiative EffectsAim to reduce uncertainty of aerosol radiative forcingAim to reduce uncertainty of aerosol radiative forcing

Investigate:

•Direct radiative effects

•Aerosol-cloud interactions (Indirect effects)

Collaboration Opportunities downwind of California

• States, FLMs, & Tribes already have Ozone and PM monitors in urban, rural, & Class I areas

• Concept to have downwind monitors coordinated during the May-June 2010 CalNEX study period

• Some external funding may be available for Ozonesondes and Radar Wind Profilers east of CA

• Funding proposal would rely upon demonstrating “in-kind” regular monitoring activities by States, FLMs, & Tribes are a significant, coordinated effort

• Primary option for external funding agency is NASA

Non-CO2 Greenhouse Gas Emissions

• What can measurements tell us?

Emissions QuantificationEmissions QuantificationImproved inventories are essential for predictive capability Improved inventories are essential for predictive capability

Bemis and Allen, 2005

Global warming potential of greenhouse gases.

Gas GWP MW

CO2 1 44CH4 23 16HFC-134a 1300 102

CO2

73%

CO2

10% Electricityimports

CH4

6%

High GWPGases 4%

N2O7%

California greenhouse gas emissions

13 May 2002

Bemis and Allen, 2005

CO2

73%

CO2

10% Electricityimports

CH4

6%

High GWPGases 4%

N2O7%

Non-CO2 Greenhouse Gas Emissions

• What can measurements tell us?2100

2000

1900

1800

1700

CH

4 (

pp

bv)

430420410400390380370

CO2 (ppmv)

CH4 (ppbv) = -888 + 7.1*CO2 (ppmv), R=.91

CH4/CO2 = 7.1 E-3 (mole/mole)GWP(CH4) = 23*16/44GWP(CH4)*CH4/CO2 = 0.06

Analysis by M. Trainer

Emissions QuantificationEmissions QuantificationImproved inventories are essential for predictive capability Improved inventories are essential for predictive capability

Global warming potential of greenhouse gases.

Gas GWP MW

CO2 1 44CH4 23 16HFC-134a 1300 102

California greenhouse gas emissions13 May 2002 WP-3D flightLos Angeles

2010: Provide more extensive data set• Spatial variability• Agricultural contribution

Bemis and Allen, 2005

CO2

73%

CO2

10% Electricityimports

CH4

6%

High GWPGases 4%

N2O7%

• What can measurements tell us?250

200

150

100

50

0

HC

F-1

34

a (

pp

tv)

430420410400390380370

CO2 (ppmv)

HCF-134a (pptv) = -1858 + 5*CO2 (ppmv), R=.92

Non-CO2 Greenhouse Gas Emissions

HCF-134a/CO2 = 5E-6 (mole/mole)GWP(HCF-134a) = 1300*102/44GWP(HCF-134a)*HCF-134a/CO2

= 0.015

Analysis by M. Trainer

Emissions QuantificationEmissions QuantificationImproved inventories are essential for predictive capability Improved inventories are essential for predictive capability

13 May 2002 WP-3D flightLos Angeles

Global warming potential of greenhouse gases.

Gas GWP MW

CO2 1 44CH4 23 16HFC-134a 1300 102

California greenhouse gas emissions

2010: • Provide more extensive data set• Investigate spatial variability• Add N2O to measurement capability• Collaborate with others doing similar

analyses with different perspectives.

Los Angeles plume advected into offshore stratus deck provides a laboratory to study aerosol indirect effect on climate

Alt

itu

de

(km

)FLEXPART transport model

GOES Visible9 June 2006

Aerosol Properties and Radiative EffectsAerosol Properties and Radiative EffectsAim to reduce uncertainty of aerosol radiative forcingAim to reduce uncertainty of aerosol radiative forcing

Analysis by M. Trainer

GOES Visible9 June 2006

Aerosol Properties and Radiative EffectsAerosol Properties and Radiative EffectsAim to reduce uncertainty of aerosol radiative forcingAim to reduce uncertainty of aerosol radiative forcing

2010: Contrast aerosol-cloud inter-actions with VOCALS results (Urban vs. smelter emissions)

Los Angeles plume advected into offshore stratus deck provides a laboratory to study aerosol indirect effect on climate

Chemical TransformationChemical TransformationAim to understand on a process level for predictive capability Aim to understand on a process level for predictive capability

Sources of Organic Aerosol in Polluted AirAt time of emission After 2 days

In 2 days:Composition changesPie shrinks: species not measured? Deposition? [de Gouw, JGR 2005]SOA formation >> calculated from known precursors

SOA growth >> expected:Semi-volatile precursors? [Robinson, Science 2007]Mechanisms poorly understood?Biogenic precursors in urban plumes?

[Volkamer, GRL 2006]

2010: • Investigate early stages of SOA

formation at central LA site.• Coordinate with all platforms

to build as complete a picture as possible

Recommended